Three hundred thirty-seven patients with primary hip or knee osteoarthritis undergoing elective lower extremity arthroplasty with hip skin biopsies were randomized to receive either standard preoperative skin preparation with 4% chlorhexidine gluconate alone (n = 173) or the same standard preparation plus four preoperative applications of 5% benzoyl peroxide gel over 48 hours and one additional application on the morning of surgery (BPO; n = 164). The primary outcome of interest was the proportion of patients with a positive dermal culture for Cutibacterium acnes from six 3-mm punch biopsies obtained after intraoperative skin preparation and before antibiotics. Secondary outcomes included the overall rate of any positive culture, the spectrum of cultured organisms, and differences in colonization between anterior versus lateral biopsy locations and between proximal, middle, and distal sites along hypothetical THA incisions. Outcomes were assessed intraoperatively at a single time point, with cultures held for 14 days. Overall, the results of the study revealed that 11% of biopsies yielded positive cultures and that C. acnes colonization rates were high and similar between groups, with no differences by surgical approach or sampling level. These findings suggest that current preoperative cleansing protocols, including the addition of BPO as used in this study, are not effective at meaningfully reducing the dermal burden of C. acnes before THA.